Study of Fibrinolytic Parameters in Different Typesof Polycythemia

G. Lugassy* and I. FilinThe Institute of Hematology, Barzilai Medical Center, Ashkelon, Israel, affiliated with the Ben Gurion University of the Negev, Beer

Sheva, Israel

Polycythemia vera (PV) is a myeloproliferative disorder characterized by thrombotic and,less often, bleeding complications. Many mechanisms have been advanced to explain theoccurrence of these complications, none of them satisfactory. We examined a cohort of27 patients with PV, secondary erythrocytosis, and essential thrombocythemia forcoagulation and fibrinolytic parameters, including euglobulin lysis test, D-dimer, anda2 antiplasmin. Ten of the 27 patients developed one or more thrombotic complica-tions during the study. We found no clinical correlation between the studied parametersand the complications. Three patients, one of each group, with elevated serum a2 anti-plasmin levels, developed severe arterial or venous thromboses. Am. J. Hematol. 60:196–199, 1999. © 1999 Wiley-Liss, Inc.

Key words: polycythemia; fibrinolysis

INTRODUCTION

Thrombosis and bleeding are frequent complicationsof polycythemia vera (PV). Several mechanisms havebeen postulated to explain these manifestations: wholeblood viscosity [1], absolute platelet count [2], plateletaggregation abnormalities [3], and decreased levels offactor XII [4].

The fibrinolytic system has been given little attentionin PV. Investigators have found decreased levels ofprekallikrein and kallikrein inhibitors [4] and reducedlevels of tissue plasminogen inhibitor antigen [5].

We conducted a prospective comparative study ofsome components of the fibrinolytic system in patientswith PV, essential thrombocythemia (ET), and secondaryerythrocytosis (SE), and looked for a possible relation-ship between the studied parameters and the develop-ment of thrombohemorrhagic complications among thesepatients.

PATIENTS AND METHODS

Diagnoses of PV and ET were established according tothe polycythemia vera study group criteria [6,7]: elevatederythrocyte mass, absence of iron deficiency, no Phila-delphia chromosome, no bone marrow fibrosis, and no

known cause for erythrocytosis or reactive thrombocyto-sis. Patients with PV, ET, or SE were either treated withphlebotomy or chemotherapy, or were untreated duringthe 12-month follow-up. None of the patients receivedantiplatelet or anticoagulant therapy at the time of test-ing. Blood counts were performed on fresh EDTA mixedblood with an STKS cell counter (Coulter Corporation,Hialeah, FL).

For coagulation studies, blood was mixed with one-tenth volume of 3.8% sodium citrate and plasma wasseparated after centrifugation at 300 rpm for 20 min at4°C. All plasma samples were stored at −70°C until used.Prothrombin time (PT), partial thromboplastin time(PTT) and fibrinogen levels were determined by an ACL1000 coagulometer (Instrumentation Laboratory, Barce-lona, Spain). For study of fibrinolysis, blood was drawnearly in the morning because of the circadian variationsof the fibrinolysis parameters. A D-dimer test was per-formed by the Latex method (Baxter Diagnostic, Dudin-gen, Switzerland).

*Correspondence to: Dr. Gilles Lugassy, Institute of Hematology, Bar-zilai Medical Center, Ashkelon, Israel.

Received for publication 5 March 1998; Accepted 4 November 1998

American Journal of Hematology 60:196–199 (1999)

© 1999 Wiley-Liss, Inc.

Euglobulin Lysis Test (ELT)

Preparation of euglobulin was as follows: the plasmaeuglobulin fraction was prepared by 20 times dilution ofcitrated plasma (0.5 ml) and acidification at pH 5.2. Afterleaving for 1 hr at 4°C followed by centrifugation(1,800xg) for 10 min at 4°C, the precipitate was dis-solved by 0.5 ml of 0.1 M Tris Hcl, pH 7.4. All thesamples were used for ELT immediately after prepara-tion.

For the ELT, 50ml of human thrombin (100 IU/ml)was placed in individual wells of a 96-well microtiterplate and the clot formation was initiated by adding 150ml of the plasma euglobulin fraction. The turbidity of thewells was measured as a function of the absorbance at340 nm every 30 min using an automatic microtiter platereader (Sigma Seiki Inc., Japan). The assay was per-formed at 37°C. The absorbance data were plottedagainst time. All the samples were assayed in duplicate.

The a2 antiplasmin level was defined by the chromo-genic method. Plasma was mixed with trisodium nitrate,3.8% in a ratio of 9 to 1, centrifuged at 1,000xg for 10min, and stored at −20°C until processed. Samples weremixed with a diluted buffer and incubated with 25-nkatplasmin diluted in 2.5 ml of plasmin dilutant. Detectionof the residual plasmin activity on thea2 antiplasminsubstrate was performed by evaluating the paranitroani-line release, monitored at 405 nm on an ACL200 coagu-lometer (Instrumentation Laboratory, Barcelona, Spain).The a2 antiplasmin levels were given in percentage ofparanitroaniline release. Results were compared withsample values obtained from 12 healthy controls. Mea-surement of plasma homocysteine was performed in se-lected cases, using a fluorescence polarization immunas-say (Abbott IMX Analyser, Abbott Park, IL). The autho-rization of the local and national Helsinki committeeswas obtained. All patients gave their informed consentprior to their inclusion in the study.

RESULTS

Twenty-seven PV, SE, and ET patients were includedin the study (Table I). Thirteen patients were diagnosedwith PV, 5 females and 8 males, with a mean age of 65years (range 47–>86 years). Eight patients were diag-nosed with SE, all men, with a mean age of 59 years(range 40–70 years). The erythrocytosis was attributed tosmoking in all patients, after other common etiologieswere excluded. Six patients were diagnosed with ET, 3females and 3 males, with a mean age of 60 years (range36–77 years). Twelve healthy controls, 6 females and 6males, with a mean age of 50 years (range 38–70 years)were also included in the study. Thrombotic complica-tions occurred in 10 of the 27 patients. Six patients hadthrombosis before the study was initiated, and 4 devel-

oped venous or arterial thrombosis during the study.None suffered from bleeding. Seven of the 10 patientswith thrombotic complications were in the PV group, 2had SE, and 1 had ET. None of the controls had throm-botic or bleeding manifestations. Hemoglobin levels anderythrocyte numbers were higher in the SE group (17.4g/dl, 5.8 × 106/mm3, respectively) than in the PV group,treated and untreated alike (16 g/dl, 5.4 × 106/mm3, re-spectively), or the ET group (13 g/dl, 4.1 × 106/mm3).Mean hemoglobin levels in the control group were 12.7g/dl. Platelet levels were higher in the ET group (mean680 × 103/mm3, range: 429–>864 × 103/mm3) than in thePV group (mean 270 × 103/mm3, range: 137–410 × 103/mm3) or the SE group (mean 210 × 103/mm3, range:167–284 × 103/mm3). In the control group, the meanplatelet count was 210 × 103/mm3, range: 186–336 ×103/mm3White blood cell counts were not significantlydifferent in all four groups. PT levels were within normallimits (above 50%) in all but 1 patient, who belonged tothe chemotherapy-treated PV group. PTT levels werenormal (below 388) in all but 5 PV patients, 1 SE patient,1 ET patient, and 1 control. Circulating anticoagulantwas not found in these patients. Values for euglobulinlysis time were comparable in PV, SE, and ET patientsand in the control group: 908 in the patients’ group vs.1008 in the control group. D-dimer levels were identicalwithin the 3 patient groups and in the control group,<0.25.

Serum a2 antiplasmin levels of the 27 patients areshown in Table I. Mean values for PV, SE, and ET pa-tients were similar to the values measured in the controlgroup: 104%, 107%, 130%, and 115%, respectively. Se-rum levels ofa2 antiplasmin were elevated (above 2 sdof the normal value) in 3 patients who had suffered se-vere episodes of thrombosis several months prior to test-ing:

1. A 63-year-old female from the hydroxyurea-treatedPV group had ana2 antiplasmin level of 167%. Shedeveloped deep vein thrombosis anderebrovscular ac-cident during the study period. The hemoglobin, he-matocrit, and platelet counts were within normal lim-its at the time of the thrombotic complications.

2. A 70-year-old male with untreated SE and a seruma2antiplasmin level of 144%, suffered a thrombosis ofthe femoral artery. His hematological profile was oth-erwise normal.

3. A 62-year-old male with untreated ET and a seruma2antiplasmin level of 174% developed a thrombosis ina digital artery. All other hematological parameterswere normal.

Plasma homocysteine levels were normal in the two lat-ter patients.

Fibrinolysis in Polycythemia: A Prospective Study 197

DISCUSSION

PV is a clonal disorder involving multipotent hemato-poietic stem cells [8]. Most individuals with PV areprone to thrombosis while few suffer from both throm-bosis and bleeding during the course of the disease [9].Furthermore, a PV patient may shift from being primarilya bleeder to being thrombosis prone as the disease pro-gresses.

Hyperviscosity due to elevated hematocrit, thrombo-cytosis, abnormal platelet aggregation, and activation ofcoagulation cascade have all been documented in PV[1–4]. No correlation has been established between thesefindings and the complications encountered in PV.

Among other possible mechanisms, the fibrinolyticsystem has been given little attention in PV. Bick [10], in1974, found subnormal fibrinolytic potential in 2 out of 4PV patients. Takahashi et al. [11] found a decreaseda2plasmin inhibitor in 9 PV patients. Recently, Wieczoreket al. [12] measured reduced plasminogen activator in-hibitor (PAI) antigen levels in PV patients who experi-enced a thrombotic event compared with normal PAIantigen levels in asymptomatic PV patients. Cohen et al.[5] confirmed the decreased PAI values in a cohort of 86PV patients, but found no correlation with the risk ofthromboembolic complications.

We examined a cohort of 27 patients with PV, SE, andET and compared their complete blood counts, D-dimerlevels, euglobulin lysis tests, and seruma2 antiplasminvalues to a population of 12 healthy controls. We lookedfor a possible correlation between several parameters offibrinolysis and the development of thrombohemorrhagiccomplications in this group of patients. Ten of the 27patients, mostly PV patients, developed at least onethrombotic episode before or during the study.

We found no abnormalities among the tested param-eters that could differentiate between asymptomatic andsymptomatic patients. Mainly, PV patients who devel-oped thrombosis had the same hematologic, coagulation,and fibrinolytic parameters than uncomplicated PV, ET,or SE patients or controls alike.

Because none of the patients suffered from hemor-rhage, the D-dimer test had little relevance in our study,since this is a more reliable test of hyperfibrinolysis thanof decreased fibrinolysis.

Elevated plasma levels ofa2 antiplasmin were foundin three patients (1 PV, 1 SE, 1 ET). These patients haddeveloped severe venous and arterial thromboses prior tothe study period. They had no evidence of chronic throm-bosis that could explain an increase ofa2 antiplasmininhibitor complexes at the time of the study.

TABLE I. a2 Antiplasmin Levels in 27 Patients With PV, SE, ET*

Diagnosis Treatment Thrombotic event Ht %Platelets

×103/mm3a2 antiplasmin

%

PV Untreated CVA, MI 57.4 204 99PV Untreated — 37.3 399 107PV Untreated Recurrent CVA 34.6 328 103PV spent Untreated — 25.2 137 55PV Phlebotomy — 48.8 237 131PV Phlebotomy MI 48 330 75PV Phlebotomy — 50.4 160 98PV Phlebotomy DVT 46.7 410 102PV Hydroxyurea — 48 494 124PV Hydroxyurea DVT 47.9 260 86PV Hydroxyurea DVT, CVA 34.9 400 167PV Hydroxyurea Recurrent CVA 42.5 199 102PV Hydroxyurea DVT 41.6 227 107SE Untreated — 51 284 89SE Untreated — 51.8 250 108SE Untreated Arterial thrombosis 47.4 272 144SE Phlebotomy — 52.6 167 128SE Phlebotomy CVA 51.9 200 100SE Phlebotomy — 51.3 256 94SE Phlebotomy — 51.6 228 94SE Phlebotomy — 52.9 190 121ET Untreated — 45.7 864 129ET Untreated Arterial thrombosis 35.7 792 174ET Hydroxyurea — 36.2 696 131ET Hydroxyurea — 38 514 120ET Hydroxyurea — 32.7 429 98ET Hydroxyurea — 44.4 504 129

*PV, polycythemia vera; SE, secondary erthrocytosis; ET, essential thrombocythemia; CVA, cerebro-vascular accident; MI, myocardial infarction; DVT, deep vein thrombosis; Ht, hematocrit.

198 Lugassy and Filin

We suggest that elevated levels ofa2 antiplasmincould be a risk factor for thrombosis. This hypothesisneeds to be confirmed in further clinical studies.

REFERENCES

1. Thomas DJ, Duboulay GH, Marshall J, Pearson TC, Ross Russell RW,Symon L, Wetherley-Mein G, Zilkha E. Effect of haematocrit on ce-rebral blood-flow in man. Lancet 1977;2:941–943.

2. Dawson AA, Ogston D. The influence of the platelet count on theincidence of thrombotic and haemorrhagic complications in polycy-thaemia vera. Postgrad Med J 1970;46:76–78.

3. Schafer AI. Bleeding and thrombosis in the myeloproliferative disor-ders. Blood 1984;64:1–12.

4. Carvallo A, Ellman L. Activation of the Coagulation System in Poly-cythemia Vera. Blood 47:669-678, 1976.

5. Cohen AM, Gelvan A, Kadouri A, Creter D, Djaldetti M. Tissueplasminogen activation levels in different types of polycythemia. EurJ Haematol 1990;45:48–51.

6. Berlin NI. The diagnosis and classification of the polycythemias. Se-min Hematol 1975;12:339–351.

7. Dudley JM, Messinezy M, Eridani S, Holland LJ, Lawrie A, Nunan T,Sawer B, Savidge GF, Pearson TC. Primary thrombocythaemia: diag-nostic criteria and a simple scoring system for positive diagnosis. Br JHaematol 1989;71:331–335.

8. Adamson JW, Fialkow PJ, Murphy S, Prchal JF, Steinmann L. Poly-cythemia vera: stem cell and probable clonal origin of the disease. NEngl J Med 1976;295:913–916.

9. Bilgrami S, Greenberg BR. Polycythemia rubra vera. Semin Oncol1995;22:307–326.

10. Bick RL. Fibrinolytic potential in polycythemia rubra vera. Beitr PathBd 1974;151:401–404.

11. Takahashi H, Hattori A, Shibata A. Profile of blood coagulation andfibrinolysis in chronic myeloproliferative disorders. J Exp Med 1982;138:71–80.

12. Wieczorek I, MacGregor IR, Prescott RJ, Ludlam CA. The fibrinolyticsystem and proteins C and S in treated polycythemia rubra vera. BloodCoagul Fibrinolysis 1992;3:823–826.

Fibrinolysis in Polycythemia: A Prospective Study 199